Construction method and system of water hammer simulation and precise protection model
Patent Information
- Application Number
- CN202410044747.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-11
- Publication Date
- 2025-07-11
Smart Images
Figure CN120296925A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of water supply pipe networks, and in particular to a method and system for constructing a water hammer simulation and precise protection model. Background Art
[0002] During the operation of pipelines for long-distance water transfer, it is sometimes necessary to cut off the water flow. If the end valve or the middle maintenance valve is closed improperly, water column separation may occur in the pipeline. The pressure after the water column separation and closure is dozens of times the normal valve closing pressure, which has great destructive power to the pipeline and threatens the safe operation of the entire pipeline system. Therefore, water column separation in the pipeline should be avoided as much as possible.
[0003] Therefore, a water hammer simulation model can be selected to investigate water flow and the propagation of pressure waves in the pipeline system, and engineering methods and improved operation strategies can be used to reduce the harm of water hammer. During the simulation process, the water hammer model needs to obtain the maximum and minimum pressure values of water hammer under the most unfavorable conditions, and also needs the parameter changes of the transient flow pressure fluctuations caused by the combined actions of different valves and pumps. It is necessary for the water hammer simulation model to accurately calculate the attenuation results of the transient flow pressure fluctuations generated by the actions of different components; however, the calculation results of most water hammer models are not accurate. Summary of the Invention
[0004] This application provides a method and system for constructing a water hammer simulation and precise protection model to solve the problem of inaccurate calculation results of the water hammer model.
[0005] On the one hand, a method for constructing a water hammer simulation and precise protection model provided by this application includes:
[0006] Construct a simulated full characteristic curve according to the pump parameters; the pump parameters include flow rate, head, rotational speed, and torque;
[0007] Construct a first model through the specific speed of the pump; the first model is a high-dimensional regression model, and the first model is configured with a higher-order function proportional to the specific speed;
[0008] Fit the parameters of the first model based on the simulated full characteristic curve to obtain the fitted candidate parameters;
[0009] Verify the candidate parameters according to a preset characteristic curve to obtain an analysis value;
[0010] If the analysis value is less than or equal to a preset threshold, output the candidate parameter associated with the analysis value as the target parameter;
[0011] Construct a water hammer simulation and precise protection model based on the first model and the target parameter.
[0012] In some possible implementations, the simulated full characteristic curve includes a first curve and a second curve. The first curve is a linear function relationship curve with the independent variables being flow rate and rotational speed and the dependent variable being head. The second curve is a linear function relationship curve with the independent variables being flow rate and rotational speed and the dependent variable being torque.
[0013] In some possible implementations, the abscissas of the first curve and the second curve are:
[0014]
[0015] The ordinate of the first curve is:
[0016]
[0017] The ordinate of the second curve is:
[0018]
[0019] Where x is in radians, h is the head parameter, β is the rotational speed parameter, v is the flow rate parameter, and m is the torque parameter.
[0020] In some possible implementations, the first model constructed by the specific speed of the water pump according to the following formula:
[0021]
[0022] WH(x)′ = a H + b H Ns + c H Ns 2 + d H Ns 3 ;
[0023] WM(x)′ = a M + b M Ns + c M Ns 2 + d M Ns 3 ;
[0024] Where N n is the rated rotational speed of the water pump, Q n is the rated flow rate of the water pump, H n is the rated head of the water pump, Ns is the specific speed of the water pump, and a, b, c, d are the parameters of the first model.
[0025] In some possible implementations, the fitting of the parameters of the first model based on the simulated full characteristic curve includes:
[0026] Substitute the first preset value into the simulated full characteristic curve to obtain a plurality of reference characteristic curves; wherein, the first preset value is the specific speed value of the first preset water pump.
[0027] Extract scatter data from the reference characteristic curves.
[0028] Use the scatter data to fit the parameters of the first model to obtain the candidate parameters after fitting.
[0029] In some possible implementation manners, the extracting scatter data from the reference characteristic curves includes:
[0030] Perform multiple linear regression analysis on the multiple reference characteristic curves, and select the data that satisfies a preset rule as the scatter data; the preset rule is an arithmetic progression rule, and the common difference is a preset value.
[0031] In some possible implementation manners, the obtaining an analysis value by verifying the candidate parameters according to a preset characteristic curve includes:
[0032] Extract first data from the preset characteristic curve; the first data is the actual head value and the actual torque value of the water pump.
[0033] Substitute the candidate parameters into the first model, and input a second preset value into the first model to obtain second data; the second preset value is the specific speed value of the second preset water pump; the second data is the predicted head value and the predicted torque value of the water pump in the first model.
[0034] Perform a comparative analysis on the first data and the second data to obtain an analysis value.
[0035] In some possible implementation manners, the method further includes:
[0036] If the analysis value is greater than a preset threshold, recalculate the parameters of the first model using the maximum likelihood estimation method to obtain the corrected parameters of the first model.
[0037] Fit the corrected parameters of the first model based on the simulated full characteristic curve.
[0038] In some possible implementation manners, constructing a water hammer simulation and precise protection model based on the first model and the target parameters includes:
[0039] Construct a preliminary selection model based on the first model and the target parameters.
[0040] Input a preset algorithm into the preliminary selection model to obtain a water hammer simulation and precise protection model.
[0041] In some possible implementation manners, the preset algorithm includes: a characteristic line algorithm, a valve closing water hammer simulation algorithm, a pump stopping water hammer simulation algorithm, a slow-closing check valve algorithm, and a water hammer simulation algorithm.
[0042] On the other hand, the present application provides a system for constructing a water hammer simulation and precise protection model, including:
[0043] Construct a simulated full characteristic curve according to the pump parameters; the pump parameters include flow rate, head, rotational speed, and torque;
[0044] Construct a first model according to the specific speed of the pump; the first model is a high-dimensional regression model, and the first model is configured with a high-order function proportional to the specific speed;
[0045] Fit the parameters of the first model based on the simulated full characteristic curve to obtain the fitted candidate parameters;
[0046] Verify the candidate parameters according to a preset characteristic curve to obtain an analysis value;
[0047] If the analysis value is less than or equal to a preset threshold, output the candidate parameter associated with the analysis value as the target parameter;
[0048] Construct a water hammer simulation and precise protection model based on the first model and the target parameter.
[0049] As can be seen from the above technical content, the present application provides a method and system for constructing a water hammer simulation and precise protection model. The method includes: constructing a simulated full characteristic curve according to the pump parameters; the pump parameters include flow rate, head, rotational speed, and torque; constructing a first model through the specific speed of the pump; the first model is a high-dimensional regression model, and the first model is configured with a high-order function proportional to the specific speed; fitting the parameters of the first model based on the simulated full characteristic curve to obtain the fitted candidate parameters; verifying the candidate parameters according to a preset characteristic curve to obtain an analysis value; if the analysis value is less than or equal to a preset threshold, output the candidate parameter associated with the analysis value as the target parameter; constructing a water hammer simulation and precise protection model based on the first model and the target parameter. Using the simulated full characteristic curve for model parameter fitting and using the preset characteristic curve not participating in the calculation to detect the accuracy of the model, a general water hammer simulation and precise protection model is obtained, meeting the accuracy requirements of water hammer simulation and the precise requirements of protection in the informatization process of the water network system. Description of the Drawings
[0050] To more clearly illustrate the technical solutions of this application, the following will briefly introduce the drawings required in the embodiments. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.
[0051] Figure 1 Flowchart of the method for constructing the water hammer simulation and precise protection model provided in some embodiments of this application;
[0052] Figure 2 Insert diagram of the function WH(x) provided in the embodiments of this application;
[0053] Figure 3 Flowchart for fitting model parameters provided in the embodiments of this application;
[0054] Figure 4 Comparison between the experimental full characteristic curve and the general model of the full characteristic curve provided in the embodiments of this application Figure 1 ;
[0055] Figure 5 Comparison between the experimental full characteristic curve and the general model of the full characteristic curve provided in the embodiments of this application Figure 2 . Specific embodiments
[0056] The following will describe the embodiments in detail, and the examples are shown in the drawings. When the following description involves the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following examples do not represent all embodiments consistent with this application. They are only examples of systems and methods consistent with some aspects of this application detailed in the claims.
[0057] To facilitate the technical solutions of the application, some concepts involved in this application will be described first below.
[0058] With the construction of water conservancy water supply projects, long-distance water diversion is becoming more and more common. During the normal operation of pipelines, sometimes it is necessary to cut off the water flow. If the end valve or the middle maintenance valve is closed improperly, water column separation will occur in the pipeline. The pressure after the water column separation and closure is dozens of times the normal valve closing pressure, which has great destructive power to the pipeline and threatens the safe operation of the entire pipeline system. Therefore, water column separation in the pipeline should be avoided as much as possible.
[0059] Therefore, a water hammer simulation model can be selected to investigate water flow and the propagation of pressure waves in a pipeline system, and engineering methods and improved operation strategies can be used to reduce the harm of water hammer. During the simulation process, the water hammer model needs to obtain the maximum and minimum pressure values of water hammer under the most unfavorable conditions, and also the parameter changes of the pressure fluctuations of transient flow caused by the combined actions of different valves, pumps and other components. It is necessary for the water hammer simulation model to accurately calculate the attenuation results of the pressure fluctuations of transient flow generated by the actions of different components; however, the calculation results of most water hammer models are not accurate. Therefore, there is an urgent need for a high-precision water hammer simulation and accurate protection model suitable for complex water networks.
[0060] Therefore, the present application provides a method and a system for constructing a water hammer simulation and accurate protection model, which uses the simulated full characteristic curve for model parameter fitting and uses the preset characteristic curve not participating in the calculation to evaluate the accuracy of the model, thereby obtaining a general water hammer simulation and accurate protection model, meeting the accuracy requirements of water hammer simulation and the accurate requirements of protection in the informatization process of the water network system.
[0061] As Figure 1 shown, the method for constructing the water hammer simulation and accurate protection model provided by some embodiments of the present application includes:
[0062] Step 1, construct a simulated full characteristic curve according to the pump parameters; the pump parameters include flow rate, head, rotational speed and torque.
[0063] First, if the flow rate Q and the rotational speed N are used as the vertical and horizontal coordinate parameters, the simulated full characteristic curve includes a set of equal H value curves (i.e., the first curve) and another set of equal M value curves (i.e., the second curve); the first curve is a linear function relationship curve with the flow rate and the rotational speed as the independent variables and the head as the dependent variable, and the second curve is a linear function relationship curve with the flow rate and the rotational speed as the independent variables and the torque as the dependent variable. Two simplified assumptions are selected as the premise for water hammer calculation using the full characteristic curve:
[0064] It is assumed that the curve measured under steady flow conditions can reflect the relationship between various parameters during the hydraulic transient process.
[0065] It is assumed that for geometrically similar pumps a and b, although their rotational speeds N and representative sizes D are different, there is a certain ratio between the parameters of similar operating conditions.
[0066] Based on the pump similarity theory, according to the similarity theory, for the same unit, D a = D b , that is, the linear ratio is λ = 1. When the rotational speed changes from N1 to N2, the following proportional relationships exist between the parameters of similar operating conditions:
[0067]
[0068] If the following new coordinate parameters (rectangular coordinate system, with the following relationships for the coordinate axes) are used to plot the curve, the characteristic curves of the same pump at various different speeds will overlap:
[0069]
[0070] Use a set of dimensionless parameters to represent the relative ratios of the operating parameters to the rated parameters, that is, convert all the pump parameters into parameters with a dimension of one:
[0071]
[0072] h is the head parameter, β is the speed parameter, v is the flow rate parameter, and m is the torque parameter.
[0073] In the formula, the values with subscript n all represent rated values. Then, the two characteristic curves in formula (7) can be redrawn as the following two characteristic curves with dimensionless parameters as the new coordinate axes (the coordinate system is the same as above):
[0074]
[0075] Using the dimensionless parameters as the coordinate axes of the new coordinate system has greater generalization ability than using the relationships in formula (7) as the coordinates of the coordinate axes. The two curves in formula (7) can represent the parameter relationships of the same pump at all speeds. And the two curves in formula (9) can be generalized to represent the parameter relationships of all pumps in all similar series at all speeds. Because when the values of β, ν, h, and m are the same, regardless of the size and speed of the similar series pumps, their operating conditions are similar.
[0076] However, since the values of β, ν, h, and m can all be positive or negative, and β may even be zero, the change ranges of the coordinate parameters v / β, h / β 2 , m / β 2 are all from -∞ to +∞, resulting in the fact that a complete curve cannot actually be drawn. Therefore, the following optimization suggestions are put forward. The dimensionless coordinate parameters in formula (9) are further transformed into the following new coordinate parameters (rectangular coordinate system with x as the horizontal axis); the simulated full characteristic curve includes a first curve and a second curve, the first curve is a linear function with the flow rate and speed as independent variables and the head as the dependent variable, and the second curve is a linear function with the flow rate and speed as independent variables and the torque as the dependent variable:
[0077] WH(x)-x, WM(x)-x
[0078] The abscissas of the first curve and the second curve are:
[0079]
[0080] The ordinate of the first curve is:
[0081]
[0082] The ordinate of the second curve is:
[0083]
[0084] The variation range of the abscissa x is from 0 to 2π, and the variation range of the ordinate will also be quite limited, so that a complete curve can be plotted in a limited coordinate system.
[0085] However, there is still a problem in the above scenario: two different operating conditions may have the same abscissa x value, causing confusion. To avoid this situation, the following regulations are proposed:
[0086] (1) ν ≤ 0, β < 0, turbine operating condition (reverse flow, reverse speed)
[0087]
[0088] (2) ν < 0, β ≥ 0, pump braking operating condition (reverse flow, forward speed)
[0089]
[0090] (3) ν ≥ 0, β ≥ 0, pump operating condition (forward flow, forward speed)
[0091]
[0092] (4) ν > 0, β < 0, reverse braking operating condition (forward flow, reverse speed)
[0093]
[0094] After the above transformation, the simulated full characteristic curve can be accurately plotted in the coordinate system.
[0095] However, the shapes of the two dimensionless simulated full characteristic curves WH(x) and WM(x) are complex and difficult to express with mathematical formulas. A method of extracting a series of discrete data from the dimensionless full characteristic curve is adopted.
[0096] A method of extracting a series of discrete data from the dimensionless full characteristic curve is as follows:
[0097] Step 1-1: Divide the interval from x = 0 to x = 2π into 88 equal parts, and the equal division interval dx = 2π / 88 = 0.0714;
[0098] Step 1-2: Extract 89 discrete values from WH(x) and WM(x) respectively, and arrange them in a table in the order of x.
[0099] When the x value obtained during the calculation process is not equal to one of the 89 discrete values, it is also necessary to determine the current WH(x) and WM(x) values through interpolation. The method is as follows:
[0100] Let Z represent the integer of (x / dx + 1), that is:
[0101] Z = INT(x / dx + 1);
[0102] As Figure 2 shown, the actual x value is between (Z - 1)dx and Zdx. The WH(I) and WM(I) values of x = (Z - 1)dx and the WH(I + 1) and WM(I + 1) values of x = Zdx can be found through the WH(x) and WM(x) tables. The curve between the two nodes is relatively small and can be approximated as a straight line. This section of the curve is approximately represented by the following formula:
[0103] WH(x) = M0 + M1x;
[0104] WM(x) = N0 + N1x;
[0105] The coefficients in the formula:
[0106]
[0107] M0 = WH(Z + 1) - ZM1dx;
[0108]
[0109] In each calculation period, in the boundary condition equation of the pump, first calculate x based on the v and β values, and then calculate the corresponding WH(x) and WM(x) values through interpolation. Combining the discrete value table and the linear interpolation method, the characteristic curve can be approximately and accurately simulated.
[0110] Step 2: After obtaining the simulated characteristic curve, construct a first model through the specific speed of the water pump; the first model is a high-dimensional regression model, and the first model is configured with a high-order function proportional to the specific speed;
[0111] In the related art, the specific speed Ns of the existing comprehensive performance curves at home and abroad is used to create a water hammer simulation model. However, due to the limited quantity of comprehensive performance curve data, when the Ns of the actually used pump is not close to the existing Ns, a large error will occur in the water hammer simulation process using the current method. Therefore, in some embodiments of the present application, by observing the WH(x)-x and WM(x)-x curves plotted from the experimental data of five pumps with specific speeds Ns = 90, 128, 260, 530, and 950, since the values of WH(x) and WM(x) change with Ns according to certain rules, a high-order function of WH and WM changing with Ns is constructed. The specific form of the model is as follows:
[0112]
[0113] WH(x)′ = a H +b H Ns + c H Ns 2 +d H Ns 3 ;
[0114] WM(x)′ = a M +b M Ns + c M Ns 2 +d M Ns 3 ;
[0115] In the formula: N n is the rated speed of the water pump, Q n is the rated flow rate of the water pump, H n is the rated head of the water pump, Ns is the specific speed of the water pump, and a, b, c, d are the parameters of the high-dimensional regression model.
[0116] Step 3: Fit the parameters of the high-dimensional regression model, i.e., the first model, based on the simulated full characteristic curve, specifically including:
[0117] Extract scatter data from the reference characteristic curve; the reference characteristic curve is the WH(x)-x and WM(x)-x curves plotted from the experimental data of four pumps with specific speeds Ns = 90, 260, 530, and 950;
[0118] Use the scatter data to fit the parameters of the first model to obtain the fitted candidate parameters.
[0119] In some embodiments, the extracting scatter data from the reference characteristic curve includes:
[0120] Perform multiple linear regression analysis on the multiple reference characteristic curves, and select the data that satisfies a preset rule as scatter data. Among them, the preset rule is an arithmetic progression rule, and the common difference is a preset value.
[0121] After constructing the first model, collect the WH(x)-x and WM(x)-x curves drawn from the experimental data of four types of pumps with specific speed Ns = 90, 260, 530, and 950, and extract the WH(x)-x and WM(x)-x scatter data of the four types of pumps using the WH(x)-x and WM(x)-x curves.
[0122] The pattern of the regression model is more obvious at specific x values, which are: 0, 0.0833π, 0.1667π, 0.2500π, 0.3333π, 0.4167π, 0.5000π, 0.5833π, 0.6667π, 0.7500π, 0.8333π, 0.9167π, 1.0000π, 1.0833π, 1.1667π, 1.2500π, 1.3333π, 1.4167π, 1.5000π, 1.5883π, 1.6667π, 1.7500π, 1.8333π, 1.9167π, 2π.
[0123] Use the scatter data of the complete characteristic curves of four known Ns values to fit the high-dimensional regression model of the complete characteristic curve, and obtain the parameters of the model at different x points.
[0124] After the above process, obtain the candidate parameters of the first model. The specific steps are as Figure 3 shown.
[0125] Step 4: Verify the candidate parameters according to the preset characteristic curve to obtain an analysis value; if the analysis value is less than or equal to the preset threshold, output the candidate parameter associated with the analysis value as the target parameter.
[0126] The preset characteristic curve is the measured complete characteristic curve of a centrifugal pump with Ns = 128. Use the measured complete characteristic curve of a centrifugal pump with Ns = 128 to test the fitted parameters.
[0127] In some embodiments, the verifying the candidate parameters according to the preset characteristic curve to obtain an analysis value includes:
[0128] Extract first data from the preset characteristic curve; the first data is the actual head value and the actual torque value of the water pump.
[0129] Substitute the candidate parameters into the first model, and input a second preset value into the first model to obtain second data; the second preset value is the specific speed value of a second preset water pump; the second data is the predicted head value and the predicted torque value of the water pump in the first model.
[0130] Compare the first data with the second data for analysis to obtain an analysis value.
[0131] Collect the WH(x)-x and WM(x)-x curves plotted from the experimental data of the pump with a specific speed Ns = 128, and extract the WH(x)-x and WM(x)-x scatter data, which is the first data. Substitute the above candidate parameters into the first model, set Ns in the model to 128, and output the second data of the first model; compare and analyze the scatter data (the first data) obtained from the experiment of the pump with Ns = 128 and the second data obtained from the first model to verify the model.
[0132] From Figure 4 and Figure 5 it can be seen that the WH(x)’-x and WM(x)’-x curves of Ns = 128 obtained from the first model proposed in this application are consistent with the experimental overall characteristic curves. The determination coefficient R of the WH(x)’-x curve 2 is 93.15%, and the determination coefficient R of the WM(x)’-x curve 2 is 98.75%. This indicates that the first model proposed in this application has the characteristic of high precision, and the water hammer simulation accuracy based on this model will be further improved.
[0133] In some embodiments, if the analysis value is greater than a preset threshold, use the maximum likelihood estimation method to recalculate the parameters of the first model to obtain the corrected parameters of the first model;
[0134] Fit the corrected parameters of the first model based on the simulated overall characteristic curve.
[0135] Step 5, construct a water hammer simulation and precise protection model based on the first model and the target parameters, which specifically includes:
[0136] Construct a preliminary model based on the first model and the target parameters;
[0137] Input a preset algorithm into the preliminary model to obtain a water hammer simulation and precise protection model.
[0138] In some embodiments, the preset algorithms include: the characteristic line algorithm, the closing valve water hammer simulation algorithm, the pump shutdown water hammer simulation algorithm, the slow-closing check valve algorithm, and the water hammer simulation algorithm.
[0139] Construct a high-precision water hammer simulation and precise protection model for complex water networks based on a preselected model, perform water hammer simulation on the complex water network, and use the high-precision water hammer simulation to combine the boundary conditions of air valves, surge towers, and air tanks to perform water hammer simulation on the complex water network. Finally, determine the location, type, and quantity of water hammer protection devices for the water network to provide precise water hammer protection for the complex water network.
[0140] Some embodiments of the present application also provide a system for constructing a water hammer simulation and precise protection model, and the system is configured to:
[0141] Construct a simulated full characteristic curve according to pump parameters; the pump parameters include flow rate, head, rotational speed, and torque;
[0142] Construct a first model according to the specific speed of the pump; the first model is a high-dimensional regression model, and the first model is configured with a high-order function proportional to the specific speed;
[0143] Fit the parameters of the first model based on the simulated full characteristic curve to obtain the fitted candidate parameters;
[0144] Verify the candidate parameters according to a preset characteristic curve to obtain an analysis value;
[0145] If the analysis value is less than or equal to a preset threshold, output the candidate parameter associated with the analysis value as the target parameter;
[0146] Construct a water hammer simulation and precise protection model based on the first model and the target parameter.
[0147] As can be seen from the above embodiments, the present application provides a method and system for constructing a water hammer simulation and precise protection model. The method includes: constructing a simulated full characteristic curve according to pump parameters; the pump parameters include flow rate, head, rotational speed, and torque; constructing a first model through the specific speed of the pump; the first model is a high-dimensional regression model, and the first model is configured with a high-order function proportional to the specific speed; fitting the parameters of the first model based on the simulated full characteristic curve to obtain the fitted candidate parameters; verifying the candidate parameters according to a preset characteristic curve to obtain an analysis value; if the analysis value is less than or equal to a preset threshold, output the candidate parameter associated with the analysis value as the target parameter; constructing a water hammer simulation and precise protection model based on the first model and the target parameter. Using the simulated full characteristic curve for model parameter fitting and using the preset characteristic curve that does not participate in the calculation to detect the accuracy of the model, a general water hammer simulation and precise protection model is obtained, meeting the accuracy requirements of water hammer simulation and the precise requirements of protection in the informatization process of the water network system.
[0148] For the similar parts between the embodiments provided in this application, reference can be made to each other. The specific embodiments provided above are only several examples under the general concept of this application and do not constitute a limitation on the protection scope of this application. For those skilled in the art, any other embodiments extended based on the solution of this application without creative efforts fall within the protection scope of this application.
Claims
1. A method for constructing a water hammer simulation and precise protection model, characterized in that, Including: Construct a simulated full characteristic curve according to the pump parameters; the pump parameters include flow rate, head, rotational speed, and torque. Construct a first model according to the specific speed of the pump; the first model is a high-dimensional regression model, and the first model is configured with a higher-order function proportional to the specific speed. Fit the parameters of the first model based on the simulated full characteristic curve to obtain the candidate parameters after fitting. Verify the candidate parameters according to the preset characteristic curve to obtain an analysis value. If the analysis value is less than or equal to the preset threshold, output the candidate parameter associated with the analysis value as the target parameter. Construct a water hammer simulation and precise protection model based on the first model and the target parameter.
2. The method for constructing a water hammer simulation and precise protection model according to claim 1, characterized in that, The simulated full characteristic curve includes a first curve and a second curve. The first curve is a linear function relationship curve with the independent variables being flow rate and rotational speed and the dependent variable being head. The second curve is a linear function relationship curve with the independent variables being flow rate and rotational speed and the dependent variable being torque.
3. The method for constructing a water hammer simulation and precise protection model according to claim 2, wherein, The abscissa of the first curve and the second curve is: The ordinate of the first curve is: The ordinate of the second curve is: Where x is in radians, h is the head parameter, β is the rotational speed parameter, v is the flow rate parameter, and m is the torque parameter.
4. The method for constructing a water hammer simulation and precise protection model according to claim 3, characterized in that The first model constructed according to the specific speed of the pump by the following formula: WH(x)' = a H + b H Ns + c H Ns 2 + d H Ns 3 ; WM(x)' = a M + b M Ns + c M Ns 2 + d M Ns 3 ; Among them, N n is the rated speed of the water pump, Q n is the rated flow rate of the water pump, H n is the rated head of the water pump, Ns is the specific speed of the water pump, and a, b, c, d are the parameters of the first model.
5. The method for constructing a water hammer simulation and precise protection model according to claim 1, wherein The fitting of the parameters of the first model based on the simulated full characteristic curve includes: Substitute a first preset value into the simulated full characteristic curve to obtain a plurality of reference characteristic curves; where the first preset value is the specific speed of a first preset pump. Extract scatter data from the reference characteristic curves. Use the scatter data to fit the parameters of the first model to obtain the candidate parameters after fitting.
6. The method for constructing a water hammer simulation and precise protection model according to claim 5, characterized in that The extracting of the scatter data from the reference characteristic curves includes: Perform multiple linear regression analysis on the plurality of reference characteristic curves, and select the data that satisfies a preset rule as the scatter data; the preset rule is an arithmetic progression rule, and the common difference is a preset value.
7. The method for constructing a water hammer simulation and precise protection model according to claim 1, wherein, The verifying of the candidate parameters according to the preset characteristic curve to obtain an analysis value includes: Extract first data from the preset characteristic curve; the first data is the actual head and actual torque of the pump. Substitute the candidate parameters into the first model, and input a second preset value into the first model to obtain second data; the second preset value is the specific speed of a second preset pump; the second data is the predicted head and predicted torque of the pump in the first model. Compare and analyze the first data and the second data to obtain an analysis value.
8. The method for constructing a water hammer simulation and precise protection model according to claim 1, characterized in that The method further includes: If the analysis value is greater than the preset threshold, recalculate the parameters of the first model using the maximum likelihood estimation method to obtain the corrected parameters of the first model. Fit the corrected parameters of the first model based on the simulated full characteristic curve.
9. The method for constructing a water hammer simulation and precise protection model according to claim 1, wherein Constructing a water hammer simulation and precise protection model based on the first model and the target parameter includes: Construct a preliminary selection model based on the first model and the target parameter. Input a preset algorithm into the preselected model to obtain a water hammer simulation and precise protection model; the preset algorithm includes: a characteristic line algorithm, a closing valve water hammer simulation algorithm, a pump shutdown water hammer simulation algorithm, a slow-closing check valve algorithm, and a water hammer simulation algorithm.
10. A water hammer simulation and precise protection model construction system, characterized in that, The system is configured to: Construct a simulated full characteristic curve according to the pump parameters; the pump parameters include flow rate, head, rotational speed, and torque; Construct a first model according to the specific speed of the pump; the first model is a high-dimensional regression model, and the first model is configured with a high-order function proportional to the specific speed; Fit the parameters of the first model based on the simulated full characteristic curve to obtain the fitted candidate parameters; Verify the candidate parameters according to the preset characteristic curve to obtain an analysis value; If the analysis value is less than or equal to the preset threshold, output the candidate parameters associated with the analysis value as target parameters; Construct a water hammer simulation and precise protection model based on the first model and the target parameters.
Citation Information
Cited By
Safety risk prediction method and system for pumping irrigation station pipeline system
CN121654900A